The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the sports-skills Python package, including instructions to fetch the source directly from the vendor's GitHub repository (github.com/machina-sports/sports-skills.git) if the standard installation is unsuccessful.
[COMMAND_EXECUTION]: The instructions guide the agent to perform environment setup and data retrieval using the sports-skills CLI and standard shell utilities like pip, which, and python3. It also includes a local validation script (scripts/validate_params.sh) that checks the system date to advise on the current NFL season status.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sports APIs, including news headlines, play-by-play descriptions, and fantasy trending data, which serves as a potential surface for indirect instructions.
Ingestion points: Data flows from external endpoints via commands like get_news, get_nflverse_play_by_play, and get_fantasy_trending.
Boundary markers: The skill instructions do not define specific delimiters or guardrails for isolating the retrieved data from the agent's instructions.
Capability inventory: The agent is authorized to execute the sports-skills CLI and write to local cache directories (e.g., ~/.cache/sports-skills/sleeper/).
Sanitization: There is no documentation of internal sanitization, filtering, or escaping of the external text content before it is processed by the agent.

nfl-data